Research On Stereotactic-assisted Localization Based On Neuronal Discharges

Stereotactic localization of brain structure based on neuronal discharges is an important task with clinical practical value. Clinic microelectrode data used in this paper from Parkinson's diseases was processed with Hilbert-Huang Transformation (HHT). The characteristic factor correlated with brain structure was explored by analyzing the characteristics of neuronal discharge signals, and the HHT characteristic factor curve of neuronal discharge signals of microelectrode piqure locus can be achieved. A method of brain structure localization was proposed, it solves location problem in neurosurgical operation and provides a reliable technology scheme for stereotactic navigation.The main work and innovations of this paper:1,Based on the characters of electrophysiological signal, such as time-dependent, nonlinear, faint and strong interference, Hilbert-Huang transformation was used to decompose the neuronal signal into intrinsic mode functions (IMF). This method realized the adaptive frequency band division of signal; it was more flexible and effective than other theories such as Fourier and wavelet analysis.2,Instantaneous frequency was introduced in the process of Empirical Mode Decomposition (EMD), and flexibility and effectiveness were increased by analyzing the signal from time and frequency domains. Several IMF components that embody the main characteristics of signal were reserved, and the high frequency noises, baseline draft and direct current component in neuronal discharge signal were eliminated.3,Applying the Hilbert transform to the main IMF components, and Hilbert marginal spectrum of signal can be obtained. The distribution rule of signal energy corresponding to different frequency can be achieved from marginal spectrums of different depths by using HHT to process the neuronal discharge signals which were recorded by the same track of microelectrode.4,Characteristic factor of neuronal discharge signals was achieved based on marginal spectrums, and localization method was proposed according to characteristic factor curve. This location method was verified with clinical data exactly.